Malignant gliomas are the most common and lethal primary central nervous system cancers and are stratified into distinct pathologic grades with well-known biologic behaviors. Although Grade I astrocytomas exhibit defined margins and may be removed surgically, most patients present with tumors that cannot be resected for anatomical reasons or at higher, more invasive stages of disease that preclude complete surgical resection. Work from many laboratories has identified mutations present in a high percentage of glioblastomas including constitutively active mutant epidermal growth factor receptor (EGFR*), inactivating mutations of PTEN, overexpression of platelet-derived growth factor (PDGF), and loss of the cyclindependent kinase 2A (CDKN2A) tumor suppressor gene. However, recent work using genome scale methods to identify regions of copy number aberrations have revealed a bewildering array of recurrent regions of amplification and deletion, presumably targeting yet-to-be-discovered glioma-relevant oncogenes and tumor suppressors. In this project we propose a systematic approach to identify oncogenes beyond those studied in Projects 1 and 2 involved in the pathogenesis of glioblastoma. We will focus on kinase and other potentially targets currently amendable to drug development in order to identify oncogenes with the highest potential for clinical translation. By combining oncogenomics and high throughput RNA interference, we will identify a set of putative oncogenes that are not only amplified in glioblastomas but also necessary for their proliferation and survival. We will then determine the roles of these candidates in tumor initiation and maintenance in functional assays in human and murine models of transformation and determine their expression patterns in patient derived glioblastoma samples. Using unbiased and pathway specific approaches, we will search for small molecule inhibitors targeting the oncogenes identified in these studies and determine the functional relationship of these oncogenes with other signature mutations in glioblastoma pathogenesis as a prelude to preclinical studies. Relevance: Despite progress in defining signature mutations in glioblastoma, the prognosis of patients with glioblastoma remains poor, and we lack curative targeted therapies for this disease. This project will identify and validate glioblastoma-relevant oncogenes, focusing on protein classes amenable,to the development of novel therapeutic agents.